Test Question Design to Disrupt Generative AI Misuse

Generative AI tools (such as ChatGPT, Claude and Gemini) don’t think or understand course concepts. They operate on probabilistic next-token prediction. Basically, they calculate the most likely sequence of words mathematically based on the patterns present in their training data. Standard textbook definitions, classic case studies and generic essay prompts are available in millions of instances online, enabling AI to generate correct answers with extremely high confidence. After all, in most cases these are the exact resources used to train the model. 

To make questions “AI-resistant”, you can write prompts and questions that force these tools into low-probability prediction territory.

5 Strategies to Break (or Bend) AI Probability Mechanics

While these strategies can help to make questions more resistant against improper use of generative AI tools, it’s important to realize that they won’t work 100% of the time. You’ll also want to carefully consider whether reworked questions are still representative of the assessment's level and aligned with the stated objectives or goals.

Quick Reference for Common Question Types

Question FormatVulnerable to AI When…AI-Resistant Adaptation
Multiple ChoiceDistractors are obvious or stem uses standard textbook phrasing.Use scenario-based stems with plausible-sounding distractors.
Multiple SelectEvaluating isolated factual statements.Require students to select all conditions that satisfy a complex, multi-constraint rule simultaneously.
MatchingMatching standard terms directly to definitions.Match the novel mini-scenarios to the applicable concept.